[Geuvadis_rna_analysis] headache

Hello, Happy new year to all of you, I hope you've enjoyed your holidays. The good news that I have to tell is that the RNAseq paper is under review; Magdalena told me this just before Christmas. So we can expect news from there in a couple of weeks. The bad news is that as I was going through some data files, I discovered a bug that is non-fatal but still needs to be fixed - and it's very early on in the process. In the processing of quantifications when I correct for the total number of mapped reads (the 45N in the file names), I've made an indexing error and corrected for the total number of unmapped reads instead. These two are very highly correlated, (see attachment; only the correlation and not the absolute numbers matter), so I'm 98% sure that none of the biological results will change. I already reran the repeat eQTL analysis for Europeans: Previously we had 5652 repeats with an eQTL, and after correction of the bug there's 5588, of which 5239 are the same as before - the difference in total numbers is 1%, and we both lose and gain some, with a total 7% change. I think this is not too bad, and I expect similar results from other reruns that I'm working on now. Permutation results were identical so I won't need to rerun them. The bug affects non-RPKM-based quantifications (exons, repeats, and LoF analyses of junctions, introns), eQTL analyses of exons and repeats, miRNA-mRNA correlations, QC analyses based on exon quantifications. Transcript and gene level analyses are based on RPKMs and are unaffected, as are miRNA quantifications. Needless to say, I'm extremely sorry and very angry at myself that I screwed up in such an early step. Most but not all of the analyses that need to be reran are mine, and deep apologies for the extra work that this will cause. If there is any analysis where you have used files with 45N, it needs to be updated. I will be rerunning things - starting from the files that affect others - during the next couple of days and will be in touch with the relevant people individually. FYI, I'm moving to San Francisco on the 8th, so being on the Pacific time zone will affect my email reaction speed and schedules of any conference calls. best, Tuuli -- Tuuli Lappalainen, PhD Department of Genetic Medicine and Development University of Geneva Medical School CMU / Rue Michel-Servet 1 1211 Geneva 4 Switzerland Tel. +41-(0)22-3795550 tuuli.lappalainen(a)unige.ch